To improve the global convergence speed of social cognitive optimization (SCO) algorithm, a hybrid social cognitive optimization (HSCO) algorithm based on elitist strategy and chaotic optimization is proposed to solve constrained nonlinear programming problems (NLPs). The proposed algorithm partitions learning agents into three groups in proportion: elite learning agents, chaotic learning agents and common learning agents. The common learning agents work in the search way of traditional SCO, chaotic learning agents search via chaotic search (CS) algorithm based on tent map which helps to avoid the premature convergence, elite learning agents search via elitist selection which helps to improve the global searching performance. Additionally, a chaotic search process is incorporated into local searching operation so as to enhance the local searching efficiency in the neighboring areas of the feasible solutions. Simulation results on a set of benchmark problems show that the proposed algorithm has high optimization efficiency, good global performance, and stable optimization outcomes for constrained NLPs.
为解决真实图像转换为动漫风格图像出现的参数量大、图像纹理和颜色损失的问题,提出了一种多通道卡通生成对抗网络(MC_CartoonGAN)。首先,使用HSCNN+(advanced CNNs for the hyperspectral reconstruction task)和遗传算法重新构建多通道图像数据集,丰富图像信息。其次,利用DenseNet网络进行特征复用减少参数的内存占用率及缓解梯度消失的问题。最后,引入多通道颜色重建损失函数,在保证了生成图像内容完整的情况下,降低了生成图像的颜色损失。实验结果表明,提出的多通道卡通生成对抗网络将真实图像转换成动漫风格图像的质量更优。
Support vector machines(SVMs)have shown remarkable success in many applications.However,the non-smooth feature of objective function is a limitation in practical application of SVMs.To overcome this disadvantage,a twice continuously differentiable piecewise-smooth function is constructed to smooth the objective function of unconstrained support vector machine(SVM),and it issues a piecewise-smooth support vector machine(PWESSVM).Comparing to the other smooth approximation functions,the smooth precision has an obvious improvement.The theoretical analysis shows PWESSVM is globally convergent.Numerical results and comparisons demonstrate the classification performance of our algorithm is better than other competitive baselines.
With a typical and simple 2-bit problem,a dynamic model of multi-agent social evolutionary algorithm(MASEA)is constructed by dynamic method.Then,the global dynamic shape of MASEA is comprehensively analyzed and the common evolution operators are also formally described.Furthermore,the effect that every evolutionary operator has on the dynamic shape is discovered by attraction analysis of the fixed points in the models.The global convergence of MASEA is also proved.
